© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 31850
    Prev
    Next
    INDEX
    Explanations

    references to traditional or accepted practices or methods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    156
    +0.27
    1.5%
    436
    +0.14
    0.8%
    92
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    246
    +0.27
    0.01
    418
    +0.14
    0.00
    436
    +0.12
    0.01
    Negative Logits
    obacterium
    -1.52
     unfolded
    -1.44
    away
    -1.42
    InstanceState
    -1.42
     ':
    -1.38
    '">
    -1.36
    '"
    -1.36
    `,`
    -1.35
    lot
    -1.34
     progressed
    -1.34
    POSITIVE LOGITS
    £
    2.73
    °
    2.45
    Īĺ
    2.32
    ¢
    2.23
    ĸ´
    2.22
    Ł
    2.21
    ´
    2.16
    ł
    2.10
    ĸ
    2.02
    ¦
    1.96
    Activations Density 0.047%

    No Known Activations